MO801 / MC972 - Topics in Architecture and Hardware (Multicore Architectures)

Notices

18/07	Notes finals released. Concept I was placed for students who do not have A on the concept page. This will last until the end of July so that you can deliver more discipline activities. On 31/07 I will release the new notes according to the activities delivered.
16/07	Notes initials released. Make sure your deliveries are ready.
07/06	Only for students who are enrolled in MC972 (undergraduate class). Course evaluation forms are available in the bin outside my door. Take a form, fill it and put it in the envelope.
16/04	List of files already handed out. Please follow the model names for the next shipments.
10/04	Deadline to define the subjects of the first and third works: 20/04.
26/03	Wiki of discipline in the air.
26/03	There will be no class on March 29. Take the time to do the $ 03 activity!
19/03	The articles are available on my IC-3 homedir, within the mo801 directory.
18/12	Check the important dates on Postgraduate calendar and graduation calendar.

Objectives

Study concepts related to Multicore Computer Architectures. The Course will be divided into two parts, one focusing on Hardware and the other focusing on Software. The software part will contain various practical activities in the laboratory.

REFERENCES

See list of articles for abstracts and class notes.

Some useful items to carry out the work:

• PTthreads: POSIX Threads Programming (tutorial)
• OpenMP: official site e tutorial

Especially in relation to the programming APIs, there are many pages with good tutorials. Do a Google search on them.

Activities

Each student must make 10 abstracts during the semester, a presentation on a specific topic and three individual papers (one theoretical and two practical).

Students enrolled only in the MC972 discipline must submit only 7 abstracts.

Abstracts

All articles below must be read and summarized during the course. Delivery dates are on the calendar at the end of this page. A summary must have only one page. As a guide, a summary should contain: Overview of the subject of the article, contribution of the article, results obtained and a comment on the relevance of this work both in relation to the area and in relation to o to the course.

1. A. Abnous and N. Bagherzadeh. Architectural Design and Analysis of a VLIW Processor. International Journal of Computers and Electrical Engineering, 21 (2): 119–142, 1995.
2. JL Cruz, A. Gonzales, M. Valero, and NP Topham. Multiple-Banked Register File Architectures. In Proceedings of the 27th Annual International Symposium on Computer Architecture, pages 316–325, Vancouver, 2000. ACM Press.
3. A. Joshi, A. Phansalkar, L. Eeckhout, and K. John. Measuring Benchmark Similarity Using Inherent Program Characteristics. IEEE Transactions on Computers, 55: 769–782, 2006.
4. NP Jouppi. Available Instruction-Level Parallelism for Superscalar and Superpipelined Machines. In Proceedings of the 3rd International Conference on Architectural Support for Programming Languages ​​and Operating Systems (ASPLOS), pages 272–282. ACM Press, 1989.
5. MS Lam and RP Wilson. Limits of Control Flow on Parallelism. In Proceedings of the 19th ISCA, pages 46–57. ACM Press, 1992.
6. MD Smith, M. Johnson, and MA Horowitz. Limits on Multiple Instruction Issue. In Proceedings of the 3rd ASPLOS, pages 290–302. ACM Press, 1989.
7. DW Wall. Limits of Instruction-Level Parallelism. In Proceedings of the 4th ASPLOS, pages 176–188. ACM Press, 1991.
8. Free: Bait Conference
9. Free: ASPLOS Conference
10. Free: Memory Wall Theme

Jobs

Each student must do the following tasks individually:

• Work 1 (theoretical): Make a reservation at wiki of the discipline.
• Work 2 (practical): The application to be parallelized is Gnumeric
• Work 3 (practical): Make a reservation at wiki of the discipline.

Presentations

Each student will make a presentation during the semester. The presentation will be on the theme of the chosen theoretical work.

Calendar

01/03	First class. Presentation and Motivation of the Course.	Presentation
06 / 03T	Creating a multicore architecture	Building a Multicore Architecture
08/03	Performance	Speedup
13 / 03T	Cache Coherence Protocols	Cache Coherence
15/03	Cache Coherence Protocols	Cache Coherence 2 (submissions)
20 / 03T	Parallel Programming Concepts	Parallel Programming Concepts (annotations)
22/03	Parallel Programming Concepts	Parallel Programming Concepts 2 (submissions)
27 / 03T	Threads	Threads (annotations)
29/03	Deadline for submission of abstracts 1, 2 and 3. There will be no class on that day.
03 / 04T	Presentation of the US $ 2.000.000,00 activity
05/04	Holiday.
10 / 04T	PThreads and OpenMP	PTthreads e Openmp (annotations)
12/04	Practical class on PThreads and OpenMP. Room 302. Exercise 1
17 / 04T	Problems and solutions when programming with threads
19/04	Class in room 302
24 / 04T	Cell processor
26/04	Cell processor. Last day for delivery of abstracts 4, 5 and 6.
01 / 05T	Holiday.
03/05	Class in room 302
08 / 05T
10/05	Class in room 302. Exercise 2
15 / 05T	Presentation (Non-conventional architectures)
17/05
22 / 05T	Presentation (Stream Processing and Use of multiprocessors in games)
24/05	Presentation (Intel Core and Core 2 processor family architecture)
29 / 05T	Presentation (Virtualization)
31/05	Presentation (Transactional Memories). Last day for delivery of abstracts 7, 8 and 9.
05 / 06T	Presentation (Java API for Multithreading Programming)
07/06	Holiday.
12 / 06T	There will be no class
14/06	There will be no class
19 / 06T	Presentation (Virtual Machines)
21/06	There was no class
26 / 06T	Presentation (AMD multicore family architecture)
28/06	Presentation (MPI - Message Passing Interface). Last day for delivery of the last summary and work 3.
10 / 07T	Exam (only for undergraduate students).